A School-Based Prevalence Estimate of Autistic Symptoms in 3-8 Year Olds: Preliminary Results From Two Indian Cities

Background:  

While estimates (based on studies in the UK and USA) suggest that India could have more than 2 million children with Autism Spectrum Disorders (ASD) (Krishnamurthy 2008), this remains to be tested. Prevalence estimates from Asian countries vary widely across time and country (China :0.003%- 0.17%, Japan: 0.011%-0.21% (Sun and Allison 2010); South Korea: 2.64% (Kim et al., 2011)). This variation could be due to a) the use of different screening and diagnostic instruments, b) differences in cultural norms for identifying ASD, and c) different levels of awareness about ASD and autistic behaviour in the target population across time and region, and d) genetic differences between populations. There is thus an urgent need to measure the prevalence of ASD in India.

Objectives:  

To conduct the first stage of a pilot epidemiological study of ASD on children in mainstream and special schools in two Indian cities, Delhi and Kolkata, using a standardized screening instrument, translated into the local languages.

Methods:  

Schools from all socio-economic sectors were selected from 3 municipal wards each in Delhi and Kolkata. A minimum of 6500 children of 3 to 8 years in each city was targeted as the initial sample size. The 12-item Social Communication Disorder Checklist (SCDC) (Skuse et al, 1995) questionnaire (sensitivity=0.9, specificity=0.69 in detecting ASD) was translated into Hindi and Bengali and validated (see Abstract#10869 Rudra et al.). The questionnaire along with information sheets and consent forms were distributed to parents/caregivers. The same questionnaire was administered personally on consenting class teachers of all schools as well as to the parents who were unable to read.

Results:  

Teacher-report data on SCDC was obtained for 7000 children in Delhi and 6500 children in Kolkata. Of these 6.3%  in Delhi and 6.74 % in Kolkata met the cut off score of 9 on SCDC. 29.74% parent response was obtained in Delhi and 51%  parent response was obtained in Kolkata. A surprisingly high percentage (Delhi: 32.32%, Kolkata :23.1%) of the children met the cut off (scored >=9) on the  parent report data on SCDC.

Conclusions:  

Preliminary data suggests a high prevalence  of autistic symptoms (measured using SCDC) on an initial sample of >6500 children each in two cities. This maybe due to a number of reasons that include a) the modest specificity of SCDC, b) over-reporting by parents who are particularly concerned about their children, and c) cultural norms and/or genetic differences at a population level. The planned next stage is to follow up the screen positive children from this stage with confirmatory diagnostic questionnaires and interviews (e.g. the ADOS and the SCQ), using a design similar to (Allison et al 2009). It is hoped that this data will contribute to generating a prevalence estimate of ASD in India that will, in turn, a) help determine the economic burden of the disease, and b) inform the development of appropriate policy for supporting children with ASD in this region.